Interview viewed 1989 times
Prof. Zweymüller, how did you come to be an orthopaedic surgeon?
I was born in Vienna, in 1941. I went to school in Baden, a little spa town near Vienna. My school was an old-style high school, where the main emphasis was on the classics. In 1959, I enrolled as an undergraduate at Vienna University, to study medicine. I qualified in 1966, and went to work as an assistant in the Department of Experimental Pathol
Prof. Zweymüller, how did you come to be an orthopaedic surgeon?
I was born in Vienna, in 1941. I went to school in Baden, a little spa town near Vienna. My school was an old-style high school, where the main emphasis was on the classics. In 1959, I enrolled as an undergraduate at Vienna University, to study medicine. I qualified in 1966, and went to work as an assistant in the Department of Experimental Pathology, while waiting for a surgical training post to become available. That was when I wrote my first paper - very "experimental", of course, - on the effects of certain drugs on the rat skeleton. Eventually, I was able to join the Department of General Surgery of the famous Vienna University Hospital, whose origins go back to the days of the Emperor Joseph II, who founded the hospital two centuries ago. Like many of my colleagues, I was sad when we had to leave the venerable old building, to move into the newly opened university hospital. In 1969, I decided to go into orthopaedic and reconstructive surgery.
Why did you want to go into medicine in the first place?
Obviously, to follow the family tradition. There were several doctors in my family, including a paediatrician, who was the Medical Director of the Vienna University Hospital, and an uncle who headed a Department of Gynaecology.
Why did you decide to become a surgeon, and, more specifically, an orthopaedic surgeon?
Medicine was too "passive" for my liking. I had, in fact, always wanted to be a surgeon. For a long time, the question was whether I should become a plastic surgeon or a gynaecologist. However, I was much influenced by Chiari, who was my teacher; and right at the start I sensed that orthopaedic surgery was going to take off in a major way and become independent of general surgery. I got my qualification as a bone surgeon in 1973.
Tell us about Chiari.
Chiari was born in Austria, and lived there all his life, until his death in 1981. He worked in the department of Albert Lorenz, the son of the great Adolf Lorenz, one of all-time greats of orthopaedic surgery in the German-speaking countries, who missed by one point getting the 1924 Nobel prize for medicine, for his work on the reduction of congenitally dislocated hips. Chiari was a great character, a fierce worker, with a knowledge of all aspects of orthopaedic surgery as practised in his day and age. It was his dearest wish to see his Department of Orthopaedic Surgery become a unit in its own right, rather than being just a part of a huge Department of Surgery. He had to fight tooth and nail to be made head of his own department within the University Hospital, and what he achieved was unique in Austria at that time. However, he was also a very quiet man, he did not travel much, and did not have a high profile outside his own country.
What sort of surgeon was he?
Like many surgeons of his time, he was not exactly fussy. We, his students, used to say jokingly that his skin incisions went right down to the bone. He was very conservative in his ideas, and anything new was immediately declared to be impossible. However, he did not close his mind entirely to these ideas, and a few months later, after careful consideration, he would come back and try out the innovation. He often told us how he had thought of the pelvic osteotomy that bears his name.
How was that?
Chiari was assisting Albert Lorenz at an acetabular shelf plasty involving a tibial graft. Lorenz did the roof osteotomy a bit too briskly, causing a fracture through the iliac bone, with major medialization of the entire distal pelvic fragment. Lorenz was devastated. Chiari was looking on, without saying anything; however, a few months later, he produced the same displacement, only deliberately.
Let's go back to your career.
Unlike Chiari, I wanted to know what the world was like outside Austria, and I travelled as much as possible during my holidays, paying out of my own pocket for all the trips abroad. This way, I went to the Mayo Clinic in August 1969, where I was allowed to assist Mark Coventry with his osteotomies.
How did you get into research on total hip replacement?
One of the first chiefs I worked for was Prof. Sletzer. He was a first-rate tumour surgeon. At that time, in the early 70s, we used cemented Charnley-Müller devices. The cementing technique was less than optimal, and our results were not great. Sletzer was casting about for a means of fixation other than cement, and we started looking at ceramics and cementless fixation. With help from the German company Rosenthal and a grant from the Department of Science, we developed an extramedullary device that ensheathed the femur and was stabilized by means of a screw, and an acetabular cup. Both components were made of ceramic material. We used the device in tumour surgery, but then, like Boutin in Pau, decided to extend its use to the management of osteoarthritis of the hip. In '73 and '74, we did studies, first in some 40 dogs, then in human cadavers; this work resulted in the development of our Metal-Ceramic Composite Endoprosthesis. This device had a stem for which cement fixation had ultimately been adopted, and a cementless ceramic acetabular component that was press-fitted into the bony acetabulum. In '75-'76, Trentani's team at the Istituto Ortopedico Rizzoli in Bologna developed a cementless stem made of ceramic-coated titanium. We used this stem in our patients. Our first results were described in a paper published in the Archives of Orthopaedic and Trauma Surgery, in 1979. Manfred Semlitsch, an engineer with the Sulzer company at Winterthur, criticized us for not having performed in-depth experimental studies before implanting the stem. So we did some trials, and the results were not good. It was found that this massive, long stem with a proximal elliptical cross-section was difficult to implant, and that there was a fairly high rate of femoral fractures. Ultimately, I thought that the only useful feature about this stem was the fact that its fixation was cementless. There were also problems on the acetabular side: The ceramic cup was only macrotextured, and tended to loosen. The situation was similar to the one found by Mittelmeier, with his ceramic threaded cup, although that device did better than ours because an unstable equilibrium tended to be set up between the ceramic and the host bone, as a result of an interposed layer of fibrous tissue. That was when we realized that a smooth implant would never become osseointegrated.
When did you start working with Semlitsch?
I had written a dissertation, in support of my application for a university teaching post, in 1978. One copy of the paper was sent to Semlitsch. I had just come back from a study trip to Britain, Canada, and the USA, when I was contacted by Semlitsch, who suggested that we should develop a cementless femoral component together with Otto Frey, who had already given some thought to this matter.
Were you, at that time, aware of the work done by Judet and by Lord?
Of course I was. However, their devices were too close to the Rizzoli stem that we had abandoned. Also, Semlitsch had convinced me that titanium was better than cobalt-chrome. This was borne out by subsequent experience. The ideas we developed together with Otto Frey involved a less massive and less aggressive stem, which was to have an rectangular rather than an elliptical cross-section to improve rotational stability. Poor resistance to torsion was still a problem with the rounded stems in use at that time.
Was that how your first cementless femoral component came about?
Yes; and when we inserted the first one into a cadaver, in June 1979, we were amazed to see the degree of primary stability achieved with this device. In fact, we had to split the femur to get the thing out again. However, the implant was in varus, like a Müller "banana"-stem, and I was very disappointed; I had lost faith in the implant. However, Otto Frey persuaded me to have another go, and we did a second cadaver study two days later. This time, the implant was perfectly aligned, and the X-ray was beautiful. So I was convinced that we had developed something that would work. We had taken the first step on the right road.
Did this femoral component allow for bone ingrowth?
No. All it had was a proximal macrotexture, and longitudinal grooves. Initially, it also had a collar. However, this feature was dropped very soon, because a collar stops the implant descending before the stem has come into contact with the cortex of the shaft. The implant is then suspended from the collar, the tip remains free, and the stem may toggle.
And what is the reason for that very pronounced trochanter wing?
This was added in order to maximize proximal support, so as to enhance both rotational and axial stability. It is a vital feature, since it compresses the bone in that proximal region. As with the rest of the stem, its contours were computed in such a way as to ensure that the lateral compressive stresses are reduced progressively rather than abruptly where the metaphyseal portion widens and the cortex becomes thinner.
You imply that the device is self-locking. Were you at all influenced by the work of Maurice Müller, who, at that time, was developing his straight-stem self-locking device?
No; and we were not going to go back to a cemented implant. We had seen that the femoral component described as "auto-bloquante" in French was usually foolproof in the sense that it could be reasonably well centred, without any major varus/valgus malpositioning of the kind that had frequently been seen with the earlier generation of thin, curved cemented implants. However, we also knew that this beneficial effect, which did much to improve the medium-term results, was obtained with only point contact with the medial and lateral cortices. The long-term results were poor, because of cement fractures around the distal portion of the stem.
Apparently, you were not at all worried about not being able to retrieve your implants?
Quite right. After all, we wanted a device that would stay in a long time, not one that would come out easily. And, indeed, we consistently found that our implants became osseointegrated. Osseointegration cannot occur unless there is perfect primary stability - also, there is a need for a microtexture, which was added in 1986.
Some manufacturers have forgone some of the osseointegration potential of their implants, by leaving polished or non-textured zones on the prostheses. It is precisely because of this ill-conceived design that many of these implants have had to be revised. We opted for an all-over ingrowth surface, and time has proved us right. Implants that cannot be removed are not really a problem nowadays, since the use of forged titanium has enabled us to get rid of the problem of stem fractures, and since the rate of infections has dropped. In other words, there is no need for retrieval nowadays.
Did you ever consider cementing this stem?
Most definitely not. Over the period from 1979 to 1989, the use of cemented stems gradually declined, and nowadays I am not doing any cemented stems at all. Our further design enhancements, based upon mathematical calculations and computer runs to optimize the geometry, have provided three-dimensional stability that allows us to dispense with cementing altogether. In our design work, we were going, from the start, for true self-locking - not just in one plane, but in three dimensions.
In what way did your concept differ from that of the other cementless stems developed in the 80s?
The essence of our philosophy was the need to achieve optimum primary stability. Trying to compress the cancellous bone will not work unless there is good bone stock around the medullary cavity. If the patient is elderly, with atrophic bone, there will be a deficiency of cancellous bone. This is why, right from the start, we were going for cortical fixation by means of a wedge effect along the edges. Our device is designed in such a way as to be inserted with a prestress. Many designers argue that this increased stress will lead to bone resorption. In actual fact, the very opposite happens: According to the law established by Wolff, who worked in Berlin in the 90s of the last century, bone that is subjected to compressive or tensile stress will develop new bone, whereas unstressed bone will be resorbed. Lintner has observed this phenomenon with joint replacements, both with femoral components and with threaded cups.
We think that the prestress given to the implant is slight, at each individual point of the device, since the total stress is spread over a surface area of several hundred square millimetres of intimate contact between the implant and the host cortex.
This physiological stress pattern is obtained by the careful preparation of the bone bed into which the implant will be impacted. The bone is not just reamed anyhow, but carefully shaped to match the geometry of the implant.
Why do you not design an "anatomical" stem, with a right and a left version of the implant?
No two patients have the same anatomical pattern, and it is vain to hope that the double bow of the proximal femur will always be perfectly matched by a so-called anatomical implant. There is only one way in which contact zones can consistently be found, and that is by working straight ahead in the longitudinal axis. The implant may be given an anteversion, but this has to be borne in mind from the start of femoral preparation. I fail to understand how, and even more so why, one would wish to fill the entire cavity. This complete fit-and-fill would destroy the endosteal blood supply in its entirety, and would thus kill the bone, preventing it from responding to stress by forming the new bone required for the osseointegration of the implant. I also think that it is vital to preserve the posterior cortex of the femoral neck. One does not see this structure on the a.p. films, especially if the femur is flared. However, one sees it at surgery and on the postoperative X-rays. Thus, the stem is fixed both proximally and distally, partly in the proximal cancellous bone, but chiefly in the cortex. We have consistently seen that an implant that is well fixed within the cortex will not subside; this observation has been particularly striking since the introduction of the SL pattern in 1986.
The reason why we can use straight, tapering implants with a rectangular cross-section is that we have devised a unique technique for the meticulous preparation of the host bed, using specially designed rasps that will allow the bone to match the implant and, thus, provide optimum compression. With curved, anatomical implants it would, at best, be difficult to control the stress distribution pattern in the bone. It may sound odd, but anatomical stems have much less intimate contact with the bone bed than does our non-anatomical design.
So why did you change the design yet again in 1986?
By that stage , we had come up against some problems. Firstly, too many patients had loosening of their cementless polyethylene cups. We had been using that cup until the end of 1984. The granulation tissue formed led to pain, which was slight at first, and which had an unusual pattern - not so much thigh pain as groin pain, as in OA patients who have not undergone hip replacement. Moreover, the proximal fixation of the first stem design left much to be desired. Each size came in two length, a short one and a long one. Sometimes, it was difficult to decide which length to choose, and the surgeon would change his mind halfway through the procedure. And there was only one neck length for all the stems, which meant that offset was insufficient for the large sizes, or excessive for the small sizes. It took us a long time to resolve these problems. The French engineer André Deckner reviewed the surgical and clinical experience obtained with the earlier stems, and wrote the computer programmes for the design of the new generation.
We spent two years talking and thinking things through, and working things out with him: The first SL stem (the SL stands for stepless) was not inserted until 28 August, 1986; it's a date that sticks in my mind, because it is my mother's birthday. We had a hard time with the manufacturer, who had to be persuaded to effect yet more modifications; however, we were determined to have a stem that was better shaped and easier to insert. We finally got an optimized range of implants, where the progression from one size to the next was no longer more or less subjective and random, but rational and controlled; what is more, the range was based upon a new geometry, and the increase in size going up through the range was exponential, with each component part growing at its own rate.
Thus, the length, width, thickness, neck length, trochanter wing, and many other features were each given their own growth factors.
The individual patients' requirements could now be satisfied much better than in the past. Another major change was the improved wedge shape, which was obtained by thickening the proximally flat stem. Once this new implant was being used, we saw a virtual disappearance of the thigh pain and subsidence problems that had been so troublesome before.
And what about your acetabular component?
After January 1985, I did not use the all-PE cup any more. At the time, I was reviewing my patients intensively, with an absolutely maniacal zeal. I looked at thousands of X-rays - and I saw very soon, from early 1984 onwards, in fact, that there was something wrong with the cup. This is why I replaced it with the pure titanium threaded cup. It was very difficult to convince Endler, Willert, and the implant firm that that was the right thing to do. So I told all the surgeons who were using the implant that the old cup should no longer be used. Some believed me; others went on using it until 1988. They now have hundreds of revision cases on their hands. However, thanks to the modifications made, with the SL stem and the titanium threaded cup, the rate of thigh pain was only 3-4%, depending on the individual surgeons. By January 1991, 100,000 arthroplasties using our device had been done, and I felt that, for the time being, the stem was in its final form. However, I found that further work had to be done to improve the threaded cup. The very heavy double thread was sometimes difficult to introduce all the way into the reamed acetabulum. Sometimes I thought I was going to rupture myself, trying to screw the cup home. As with the stem, it took time to effect the required improvements. If an implant is doing well, the manufacturers will be very reluctant to accept new and costly modifications, which will also, to some extent, affect the credibility of the device and interfere with the follow-up. However, I would not listen to any such arguments. In 1991, major changes took place in the company I had been working with up until then. All the original team left; and I did not get on with the new management. So I left the company in 1992, to work again with my former colleagues, but in a new company. It is with them that I have developed my new biconical cup, which has worked so well and which has so much extended the range of indications, especially in revision surgery. Also, the SL stem is being enhanced further.
You mentioned revision surgery. What is your attitude nowadays?
This is the real challenge for every hip surgeon. In revision arthroplasty, my priority is to obtain implant fixation in living host bone. I do not like a fixation that is mainly obtained by means of a bulk allograft, for instance a femoral head from the bone bank; such allografts are just chunks of dead bone. I use allografts only for the filling in of defects, not in order to provide stability. If my threaded cup cannot be properly stabilized in this way, I use a cemented Burch-Schneider reinforcement device, which, to me, is the most elegant solution. Incidentally, since I started using my new cup, I have been using fewer and fewer Burch-Schneider devices. If I cannot use a Burch-Schneider, I do a Girdlestone. I have used some saddle prostheses, but the results are not as satisfactory as following a good Girdlestone: Flexion is at most 60°; the patients have to go on using crutches; also, there is a high rate of infection.
After 18 years' development, do you think that your product is now in its final form?
Yes. To me, cementless fixation is no longer a problem. Over the last five years, I have personally done more than 600 primary arthroplasties, without any cement for the femoral component, and using cement for the fixation of a cup in a Burch-Schneider reinforcement only once, in a patient with severe osteoporosis. I do not have an age limit for my cementless implant patients, and the results obtained are identical in all respects, in all the patients treated. The final problem that causes concern is osteolysis. We know that PE debris that gets into the joint is a very bad thing. Since 1979, I have been using ceramic heads. However, even though the ceramic-on-PE combination is better than metal-on-PE, I know that this is not the ultimate solution to the problem.
Would you propose to solve the problem with a metal-on-metal combination?
Yes. The McKee-Farrar prosthesis was not always perfectly machined; however, with that device, metal synovitis was extremely rare, even after many years in situ. The main problem with the McKee-Farrar was not the metal-on-metal combination, except where the components were poorly matched; it was the cemented fixation of the implant. I remember revising some McKee-Farrars myself, for catastrophic loosening after 16-17 years' in situ; even with this loosening, however, there was not the slightest metal debris synovitis. The surfaces looked like new. This, to me, was convincing - the more so, since present-day metal-on-metal combinations are very sophisticated.
Are you not worried about lymphomas and about other cancer risks described by some authors?
No; but this is still a very controversial subject. So far, the circulating cobalt levels that we have observed have been extremely low. Of course, there is a real problem of metal surface deterioration when there is subluxation or dislocation of the joint; or if the surgeon is not meticulous in his handling of the metal implants. In actual fact, the chief question for us is which of the combinations available - Sikomet or Metasul - will be the more efficient in the very long run. I do not wish to comment further on this, at this point in time; suffice it to say that I am very optimistic.
Are you considering any other developments, especially in primary arthroplasty?
Apart from the development of anti-dislocation inserts for a metal-on-metal combination, with which I am experimenting at the moment, I think that I shall have to wait until we shall have followed up for a longer time what we are doing now, before thinking of further changes. My surgical work is totally routine, at the moment. Surgery is quick; primary stability is obtained at once; and this allows me to try out all the combinations and permutations: mini-stems; ceramic-on-PE or metal-on-metal combinations; five neck lengths for the metal heads; large diameter ceramic heads; etc.
So, it's really "wait and see"?
Why did you give up your university post?
My position in the Department of Orthopaedics of the University Hospital was great; unfortunately, though, if one is not right at the top of the hierarchy, there are limits to what one can do. I was able to take up the medical directorship of an orthopaedic hospital in Vienna, the Gersthof Hospital. In this new job, I have been free to try out new developments, and I have been entirely my own master in anything to do with my research.